In HIV-infected subjects taking HAART, in whom HIV replication has been substantially suppressed, HIV infection is not eradicated. Viral latency occurs when HIV DNA becomes integrated into the genome of human cells during the cycle of infection, and then becomes transcriptionally silent (or nearly so). There is a small cache of resting T lymphocytes in the body that contains non-transcribed HIV DNA (or very slowly transcribed) that is integrated into the human genome. Without transcription of HIV genes these infected T cells presumably do not present foreign antigen. They are immunologically inert, their HIV DNA becomes transcriptionally silenced, and the infection is latent. The turning over of T cells that contain latent HIV infection is not rapid enough to achieve their elimination naturally over time while taking HAART. Based primarily upon data gathered from sampling lymphocytes in blood plasma, it is widely suggested that the eradication of HIV infection could be achieved if new measures are taken to eliminate cells that contain latent HIV, especially the small proportion of T cells that harbor HIV latency. To find cellular caches of latent HIV in the body and eliminate them challenges must be faced: 1) Do changes in lymphocytes sampled from blood plasma samples obtained clinically reflect what occurs in deep body compartments, especially non-lymphoid organs including the central nervous system (CNS), in which infection is primarily in macrophages, microglia and astrocytes? 2) Is there a macrophage/microglial phenotype in the CNS and/or tissue histiocytes of deep organ compartments that specifically support HIVlat that should be selectively targeted for elimination? 3) Are the cellular caches of HIVint in the CNS and other deep body compartments distributed homogeneously, and in characteristic cell types? Of special concern is that latency in mononuclear phagocytes, which includes the widespread tissue histiocytes of the body, is not understood. These are the cells that are critical reservoirs in the central nervous system (CNS) and other non-lymphoid types of tissue. These basic questions are difficult to address clinically because access to human tissue specimens is necessary. This program of study will utilize tissue specimens from well characterized subject who were infected with HIV to define the cellular caches of HIV DNA that has been integrated into the human genome, and thus, could contribute to HIV in the body latency. One aim will define this pool of HIV DNA in the CNS. A second aim will produce similar data in other deep tissue compartments of the human body. These data will assist the field in targeting cells that harbor latent virus so that they can be eradicated.